Increase tolerance for a flaky test (test_single_return_value) (#703)

* update dev version

* increase tolerance

* update changelog

* trigger CIs

* Auto update version from '0.36.0-dev44' to '0.36.0-dev45'

---------

Co-authored-by: ringo-but-quantum <github-ringo-but-quantum@xanadu.ai>

.github/CHANGELOG.md

@@ -132,6 +132,9 @@
 * Update the version of `codecov-action` to v4 and fix the CodeCov issue with the PL-Lightning check-compatibility actions.
   [(#682)](https://github.com/PennyLaneAI/pennylane-lightning/pull/682)
 
+* Increase tolerance for a flaky test.
+  [(#703)](https://github.com/PennyLaneAI/pennylane-lightning/pull/703)
+
 ### Contributors
 
 This release contains contributions from (in alphabetical order):

pennylane_lightning/core/_version.py

@@ -16,4 +16,4 @@
    Version number (major.minor.patch[-label])
 """
 
-__version__ = "0.36.0-dev44"
+__version__ = "0.36.0-dev45"

tests/lightning_qubit/test_measurements_class.py

@@ -474,7 +474,7 @@ def test_single_return_value(self, measurement, observable, lightning_sv, tol):
         result = m.measure_final_state(tape)
 
         # a few tests may fail in single precision, and hence we increase the tolerance
-        assert np.allclose(result, expected, max(tol, 1.0e-5))
+        assert np.allclose(result, expected, max(tol, 1.0e-4))
 
     @flaky(max_runs=5)
     @pytest.mark.parametrize("shots", [None, 1000000])